The effectiveness of low level laser therapy and acupuncture as interventions for ...

38 Pages·2016·0.48 MB·English

by  Abel Getachew

The Plymouth Student Scientist, 2016, 9, (1), 24-61

The effectiveness of low level laser therapy
and acupuncture as interventions for
temporomandibular joint disorders in adults: a
systematic review and meta-analysis

Abel Getachew

Project Advisor: Stephen Thompson. School of Biomedical and Healthcare Sciences
(Plymouth University Peninsula Schools of Medicine and Dentistry). Drake Circus,
Plymouth, PL4 8AA

Abstract
Temporomandibular joint disorders (TMD) encompass a range of disorders of the
temporomandibular joint, the masticatory muscles and other associated structures. The main
symptom of TMD patients is pain within the orofacial region. The objective of this review is to
compare the effectiveness of low level laser therapy (LLLT) and acupuncture as
interventions for TMD. Randomized controlled trials comparing LLLT versus PLT and real
acupuncture versus placebo acupuncture were included within the review. The primary
outcome was subjective pain intensity expressed via a numerical visual analogue scale
(VAS) upon palpation of the masseter muscles. Secondary outcomes include pain intensity
via VAS upon palpation of other areas of the myofascial region; the lateral pole of the
condyle, the pre-auricular region and the external auditive meatus. The author performed the
data extraction, analysis and the risk of bias assessment. 10 studies (n=317) were included
in assessment of LLLT vs PLT. LLLT was found to be statistically more effective than PLT in
reducing subjective pain intensity upon palpation. Six studies (n=165) were included in the
assessment of real acupuncture versus placebo acupuncture. Acupuncture was not
statistically more effective in reducing subjective pain compared to placebo acupuncture in
TMD patients. The I2 statistic described the percentage of variability in the effect estimates
from the different subgroups which shows considerable heterogeneity across the subgroups.
In comparing both treatments as measures for managing pain intensity in patients with TMD,
LLLT significantly reduced subjective pain on palpation of the masseter muscles, lateral pole
of the condyle, the pre-auricular region and the external auditive meatus. Acupuncture
therapy, on the other hand, did not significantly reduce pain intensity upon palpation of the
masseter muscles in patients with temporomandibular disorders. The results suggest that
LLLT is a more effective non-invasive intervention for TMD.

[24]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Table of Contents
Abstract .............................................................................................................................................. 24
Background ....................................................................................................................................... 27
Description of the condition .................................................................................................. 27
Description of the intervention ............................................................................................. 27
Why it is important to do this review .................................................................................. 28
Objectives .......................................................................................................................................... 29
Methods .............................................................................................................................................. 29
Types of studies ....................................................................................................................... 29
Types of participants .............................................................................................................. 29
Types of outcome measures ................................................................................................. 29
Search methods for identification of studies ................................................................... 29
Data collection and analysis ................................................................................................. 30
Selection of studies ................................................................................................................. 30
Data extraction and management ........................................................................................ 30
Assessment of risk of bias in included studies ............................................................... 30
Measures of treatment effect ................................................................................................ 30
Subgroup analysis and investigation of heterogeneity ................................................. 30
Results ................................................................................................................................................ 31
Description of studies ............................................................................................................. 31
Results of the search .............................................................................................................. 31
Included studies ....................................................................................................................... 31
Excluded studies ...................................................................................................................... 31
Risk of bias in included studies ........................................................................................... 32
Allocation (selection bias) ..................................................................................................... 32
Blinding (performance bias and detection bias) ............................................................. 32
Incomplete outcome data (attrition bias) ........................................................................... 32
Effects of interventions .......................................................................................................... 32
Discussion ......................................................................................................................................... 36
Summary of main results ....................................................................................................... 36
Overall completeness and applicability of evidence ...................................................... 37
Quality of the evidence ........................................................................................................... 37
Potential biases in the review process ............................................................................... 37
Agreements and disagreements with other studies or reviews .................................. 37
[25]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Authors' conclusions ...................................................................................................................... 38
Implications for practice ......................................................................................................... 38
Implications for research ....................................................................................................... 38
Acknowledgements ......................................................................................................................... 38
Characteristics of studies .............................................................................................................. 39
Characteristics of included studies .................................................................................... 39
Characteristics of excluded studies .................................................................................... 55
References to studies ..................................................................................................................... 56
Included studies ....................................................................................................................... 56
Excluded studies ...................................................................................................................... 57
Other references............................................................................................................................... 58
Data and analyses ............................................................................................................................ 61

[26]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Background

Description of the condition
Temporomandibular disorders (TMD) encompass a wide range of disorders of the
temporomandibular joint (TMJ), the masticatory muscles as well as their associated
structures and has been identified as the leading cause of pain in the orofacial region
excluding dental pain (Herranz-Aparicio et al. 2013). The vast majority of patients
with TMD also present symptoms of pain dysfunction syndrome which can be a
combination of any of the following signs and symptoms: pain on palpation of the
TMJ, pain on palpation of associated muscles, restriction or deviation of mandibular
movement, joint sounds and headaches (Gray et al. 2003). The main symptom
expressed by patients suffering from TMD is pain within the masticatory muscles
which can have a debilitating effect on patient's lives (de Moraes Maia et al. 2011).
Up to 33% of the population may experience TMD within their lifetime which can
have serious implications within a person's life (Wright and North 2009). The two
matching temporomandibular joints on either side of the skull located just in front of
the ears are the source of pain in temporomandibular joint disorder patients.

Many of the symptoms of TMD are caused by physical stress of the cartilage,
muscles, nearby ligaments as well as the teeth. For numerous patients the cause is
often not known but may be due to an improper lining of the teeth, grinding of teeth
at night (bruxism) or even poor posture (A.D.A.M 2014). Common pathologies that
cause pain within the TMJ include disc displacement and degenerative joint diseases
such as arthritis and arthrosis which is supported by evidence of the joint being a
very heavily loaded structure (Cairns 2010). Extreme loading of the joint has been
found to initiate peripheral mechanisms which cause pain by the mechanical
stimulation of nociceptors and an up-regulated release of substance P and calcitonin
gene-related peptide (CGRP) (two neuropeptides) and proinflammatory cytokines
such as tumour necrosis factor alpha (TNFα) and interleukins 6 and 8 (Cairns 2010).
The inflamed joint has been also been found to increase the nociceptive input which
is effective in prompting central sensitization which may contribute to an increased
association of inflammation within the TMJ causing pain (Yu et al. 1996; Ohrbach
2010).
Description of the intervention
The theory of using LLLT was first introduced by Mester and colleagues in 1968
(Mester et al. 1968). Since then, there has been considerable development in the
clinical applications of this therapy as is evident in its application to a wide range of
disorders, from bone healing to pain reduction (Rola et al. 2014). The use of lasers in
treating patients has continuously improved since the late 60's and is now a very
common practice within dentistry to relieve pain and inflammation. Lasers with a
higher output can damage and even destroy cells thus with low-level lasers, the non-
thermal therapy can promote tissue and cellular modifications through several
different kinds of metabolic pathways (de Moraes Maia et al. 2011). Examples
include an increased activity within the mitochondria, increasing amounts of
vascularization and the synthesis of fibroblasts which can all aid in tissue healing
and remodelling (de Moraes Maia et al. 2011). Wound healing comprises of a very
intricate interaction between several cell types and can be divided into three phases;
an inflammatory phase, a proliferative phase and a remodelling phase. Fibroblasts,
[27]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

which are key players within the proliferative phase, have been extensively studied in
regards to the effect of LLLT on their growth and locomotion (Chung et al. 2011;
Posten et al. 2005).

The output range of low level lasers is between 1-1000mW and at a wavelength
between 632 and 1064 nm a biological response is elicited. The lasers are safe to
use since they do not emit heat, vibration or sound and acts by stimulation of a
photochemical reaction within cells which is commonly referred to as biostimulation
or photobiomodulation (Hashmi et al. 2010). When an electron within a treated
chromophore absorbs photons of light the cell becomes excited and jumps from a
low-energy orbit to a higher one (Sutherland 2002) which consequently stores
energy which can then be used to carry out a range of cellular functions (Chung et
al. 2011). Other cells such as immune cells have also been found to be greatly
affected by LLLT. For example, mast cells which are very important for the motility of
leukocytes have been shown to be degranulated when in contact with specific
wavelengths of light which consequently results in the excretion of TNFα leading to
an increased infiltration of leukocytes within tissues (el Sayed and Dyson 1996;
Chung et al. 2011).

Acupuncture is a process the process of insertion and stimulation of needles into
specific areas of the body aiding health regeneration (Vickers et al. 2013). This
contemporary medicinal therapy is thought to have originated from China centuries
ago; however, this therapy has dealt with substantial controversy over the years due
to poorly defined cellular and biological mechanisms detailing how this therapy can
relieve pain (Takano et al. 2013). Acupuncture therapy is one of the most common
non-pharmacological analgesics used in treating a wide variety of pain syndromes
from Bell's palsy to tennis elbow. One theory which explains acupuncture-mediated
analgesia is the release of opioid peptides within the central nervous system
(Takano et al. 2013). Most experts within the field have agreed that acupuncture
stimulates the release of endogenous opiates such as β-endorphin and endomorphin
which subsequently activates the μ- and δ-opioid receptors; however, studies have
also looked into the role of serotonin (Lin 2008). The association of endogenous
opiates and the serotoninergic descending inhibitory pathway has also been
suggested to be a crucial mechanism of acupuncture analgesia (Lin 2008). During
acupuncture therapy, Aβ, Aδ and C afferent fibres are all activated and when the Aβ
and Aδ fibres are excited enough this results in the induction of an analgesic effect
(Zhao 2008). Acupuncture "trigger points" have been found to be stimulated by heat,
electrical currents, pressure and laser light as well as shock waves resulting in
several methods of conducting this therapy (Ernst 2006). For this review, studies
which only used the conventional acupuncture needles were considered.
Why it is important to do this review
To date there are no systematic reviews which compare the effectiveness of LLLT
and acupuncture in a head-to-head comparison. This is mainly due to a lack of high
quality clinical trials comparing both treatments. Due to the absence of comparative
reviews examining both treatments, sub-group analyses can provide details of the
effectiveness of each treatment as an intervention for temporomandibular joint
disorders.

[28]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Objectives
To compare the effectiveness of low level laser therapy and acupuncture as
interventions for treating pain caused by temporomandibular joint disorders.

Methods

Types of studies
Randomized, placebo-controlled clinical trials of low level laser therapy versus
placebo laser and real acupuncture versus placebo (sham) acupuncture as
interventions in treating temporomandibular joint disorders were included. Quasi-
randomized trials were excluded from this review.

Types of participants
Adults aged equal to or over 18 years who presented with temporomandibular joint
disorders according to the Research Diagnostic Criteria for Temporomandibular
Disorders (RDC/TMD) guidelines met the inclusion criteria. Those who were
assessed by either complete clinical examinations of chronic pain levels (at least 4
times per week over a 12 week period) of the temporomandibular joint, have
diagnosed myofascial pain syndromes, suffer from temporomandibular
symptomatology such as pain, joint sounds or osteoarthritis or present established
myofascial trigger points were also included within this review.

Types of interventions
For low level laser therapy trials, the experimental intervention was an active laser
whereas for the placebo intervention a placebo laser producing no output was used
as a control. For acupuncture trials, the experimental intervention was real
acupuncture as opposed to placebo (sham) acupuncture as a control. The control
acupuncture procedure did not involve penetration of the skin though a blunt needle
was used through a foam pad to prevent patients from knowing which treatment they
had received. Studies which had used a laser to provide acupuncture as a treatment
were excluded.
Types of outcome measures
Primary outcomes
The primary outcome measure used was a visual analogue scale (VAS) which is a
numerical scale used to measure subjective pain on the masseter muscles.
Secondary outcomes
Secondary outcomes were also pain intensity via VAS however upon palpation of
other areas of the myofascial region; the lateral pole of the condyle, the pre-auricular
region and the external auditive meatus.
Search methods for identification of studies
To identify studies to be included within the review, detailed search strategies were
used for the following databases which had no language or date restrictions:
 MEDLINE/PUBMED Central (whole database to 20th of January 2015)
 The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane
Library 2014, Issue 12)
 Web of Science (whole database to 21st of January 2015)
[29]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Data collection and analysis
The author collected and analysed all the data presented within this comparative
review.

Selection of studies
One review author independently assessed the abstracts of each relevant study from
the searches performed. Upon completion of the abstract assessment, relevant
studies were chosen and full articles were obtained for further assessment.

Data extraction and management
Raw data was extracted from each included trial into a workbook which was
subsequently collated and organized prior to input into the statistical analysis
software (RevMan 5.3) which was used.

Assessment of risk of bias in included studies
To assess the risk of bias of each study, the assessment tool in RevMan 5.3 was
used. The areas assessed for bias were selection, performance, detection, attrition
and reporting bias. To assess whether there was selection bias, adequate
information regarding the method of random sequence generation and allocation of
treatments to the patients had to be provided. Performance bias included the
blinding of participants and personnel involved within the study. Detection bias
entailed blinding of the outcome assessor and attrition bias refers to the reporting of
incomplete data from the study. Finally, the reporting bias included adequate
selective reporting of the data. The results from the assessment of bias is
summarised in figure 1 and 2.
Measures of treatment effect
The primary outcomes measured were subjective pain intensity via a VAS on
palpation of the masseter muscles. The secondary outcomes were also a measure
of subjective pain intensity on palpation of the lateral poles of the condyle, pre-
auricular regions and the external auditive meatus also rated on a VAS.

Dealing with missing data
Two studies (Smith 2006 and Mazzetto 2010) were not included within the analysis
due to missing raw data from their studies.

Assessment of heterogeneity
By reviewing the included studies characteristics, heterogeneity was assessed.
Heterogeneity was also assessed by reviewing the forest plots presented (see
Figure 3-7) in particular the Tau2 and Chi2 values as well as the confidence interval
and the I2 statistic.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis was performed for low level laser therapy studies. Heterogeneity
within these studies was assessed by referring to the characteristics of the included
studies as well as examining the forest plots (see Figure 1.2 - 1.4). The
I2 percentage of each meta-analysis indicates that there is a considerable degree of
heterogeneity across the included studies. An indirect comparison was conducted
between the interventions due to the absence of head-to-head randomized
controlled trials. The difference between the summary effects within the sub-groups
provides an estimate in comparing the two treatments together (LLLT and
[30]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Acupuncture). The validity of this indirect comparison relied heavily on the different
sub-groups of trials being similar, on average, in all aspects which may affect the
outcome. Essentially, an observational finding was conducted across the trials which
are very liable to bias.

Results
Description of studies
See: Characteristics of included studies and Characteristics of excluded studies for
detailed information.

Results of the search
16 potentially eligible studies were found; 10 studies (n = 211) for LLLT and 6 studies
(n = 165) for acupuncture.

Included studies
For low level laser therapy, 10 randomized clinical trials (RCTs) (n=347) met the
inclusion criteria (Carrasco 2008; Conti 1997; da Cunha 2008; da Silva 2012;
Emshoff 2008; Kulekcioglu 2003; Madani 2014; Mazzetto 2007; Venancio 2005 and
Venezian 2010). Within the 10 included studies, 8 studies assessed the
effectiveness of LLLT on palpation of the masseter muscles via a Visual Analogue
Scale (VAS). The remaining 2 studies (Carrasco 2008 and Mazzetto 2007) as well
as da Silva 2012 also examined the effectiveness of LLLT but on different areas of
the orofacial region; the lateral pole of the condyle, the pre-auricular region and the
external auditive meatus which was included as a sub-group analysis. The sub-
group analyses also had an outcome of subjective pain on palpation presented on a
VAS.

For acupuncture therapy, six randomized clinical trials (n=165) met the inclusion
criteria (Diracoglu 2012; Goddard 2002; Itoh 2012; Shen 2007 and 2009 and Tekin
2013). All six clinical trials assessed the effectiveness of acupuncture therapy versus
placebo (sham) acupuncture on relieving subjective pain of the TMJ upon palpation
of the masseter muscles expressed on a VAS scale. For all placebo acupuncture
therapies, the skin was lightly pricked with a blunted acupuncture needle through a
foam pad without penetrating the skin. Diracoglu 2012 also assessed the use of
acupuncture on the pain pressure threshold (PPT) with a pressure algometer and
measurements of unassisted jaw opening without pain. Itoh 2012 additionally
assessed oral function by means of measuring the maximal mouth opening possible
without pain.

More details of each trial can be found in the Characteristics of included studies
tables.

Excluded studies
Eight studies did not meet the inclusion criteria and were excluded; details on
reasons can be found in the Characteristics of excluded studies table. The two main
reasons for exclusion within these eight trials were due to the inability to access the
raw data for statistical analysis. Other reasons include the study design not being a
randomized controlled trial.

[31]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Risk of bias in included studies
The risk of bias of each study was assessed by one author. Figure 1 and 2 displays
a summary assessment and the risk of bias tables, respectively, adapted from the
Characteristics of included studies.

Allocation (selection bias)
In 12 studies (Carrasco 2008; da Cunha 2008; da Silva 2012; Diracoglu 2012;
Goddard 2002; Itoh 2012; Kulekcioglu 2003; Madani 2004; Shen 2007; Shen 2009;
Tekin 2013; Venancio 2005) there was insufficient detail provided to justify either a
‘high risk’ or ‘low risk’ judgement. Consequently, an ‘unclear risk’ was allocated to
each of these studies. The remaining 4 studies provided sufficient detail to warrant a
low risk of bias.

Blinding (performance bias and detection bias)
All studies achieved a ‘low risk’ of bias in regards to the blinding of participants and
personnel due to sufficient detail provided on the methods of blinding. Two studies
(Kulekcioglu 2003; Venancio 2005) received a ‘high risk’ of bias in the blinding of the
outcome assessments. Mazzetto 2007 did not provide sufficient information on the
blinding of the outcome assessor to warrant a ‘low risk’ of bias consequently
awarding this study an ‘unclear risk’ for the blinding of the outcome assessments.

Incomplete outcome data (attrition bias)
Shen 2007 did not provide any information regarding any drop-outs or missing
outcome data suggesting there may be a ‘high risk’ of bias. Conversely, all other
studies provided sufficient information on any missing data or drop-outs from the
trials with appropriate explanations.

Effects of interventions

Low Level Laser Therapy versus Placebo Laser Therapy

VAS of masseter muscles
Eight studies (Conti 1997; da Cunha 2008; da Silva 2012; Emshoff 2008; Kulekcioglu
2003; Madani 2014; Venancio 2005 and Venezian 2010) (n=241) were included in
this comparison whose outcome was assessed by a visual analogue scale (VAS). In
favour of low level laser therapy, a statistically significant difference was found
(standardized mean difference (SMD) = -0.29; 95% confidence interval (CI) -0.55 to -
0.02, P = 0.03) suggesting that active laser therapy is more effective than placebo
laser therapy in reducing subjective pain on palpation of the masseter muscles

(Figure 3, Analysis 1.1).

[32]
The Plymouth Student Scientist, 2016, 9, (1), 24-61

Figure 1: Risk of bias summary: review authors' judgements about each risk of bias item for
each included study.

Figure 2: Risk of bias graph: review authors' judgements about each risk of bias item
presented as percentages across all included studies.

[33]